most citedA kinetic-based regularization method for data science applications

1 citations · 2 across the 5 of their papers we have counts for

collaborators

5 papers

math.NA2026

Kinetic-based regularization: Learning spatial derivatives and PDE applications

Abhisek Ganguly, Santosh Ansumali, Sauro Succi

Accurate estimation of spatial derivatives from discrete and noisy data is central to scientific machine learning and numerical solutions of PDEs. We extend kinetic-based regulariz…

cs.LG2026

Deep Neural Networks as Discrete Dynamical Systems: Implications for Physics-Informed Learning

Abhisek Ganguly, Santosh Ansumali, Sauro Succi

We revisit the analogy between feed-forward deep neural networks (DNNs) and discrete dynamical systems derived from neural integral equations and their corresponding partial differ…

cs.AI2025

Dual Computational Horizons: Incompleteness and Unpredictability in Intelligent Systems

Abhisek Ganguly

We formalize two independent computational limitations that constrain algorithmic intelligence: formal incompleteness and dynamical unpredictability. The former limits the deductiv…

cs.LG2025★ 1 cited

Randomness and signal propagation in physics-informed neural networks (PINNs): A neural PDE perspective

Jean-Michel Tucny, Abhisek Ganguly, Santosh Ansumali +1

Physics-informed neural networks (PINNs) often exhibit weight matrices that appear statistically random after training, yet their implications for signal propagation and stability…

cs.LG2025★ 1 cited

A kinetic-based regularization method for data science applications

Abhisek Ganguly, Alessandro Gabbana, Vybhav Rao +2

We propose a physics-based regularization technique for function learning, inspired by statistical mechanics. By drawing an analogy between optimizing the parameters of an interpol…